- Email: [email protected]
Body water estimates in neonatal polycythemia
January 1983 The Journal o f P E D I A T R I C S
113
Body water estimates in neonatal polycythemia To determine whether neonatal polycythemia and its treatment by partial exchange transfusion affect body water estimates, 10 normocythemie and eight polycythemic neonates were studied within 12 hours of birth. Total body water, extracellular water, and plasma volume were estimated immediately prior to and following exchange, lntracellular and interstitial water contenr were calculated. There were no significant differences between normocythemic and preexchange polycythemic neonates in mean total body water, extracellular water, interstitial water, and intracellular water contents. In the polycythemic group, exchange did not affect mean total body water, but was associated with decreases in mean extracellular water and mean interstitial water and an increase in mean intraeellular water. Mean transeapillary escape rate of T-1824 was not affected by exchange but was quite rapid both before (35 + SE 3%/hr) and after the procedure (30 +_ 4.9%/hr). These data suggest that moderate polyeythemia in normal term neonates does not affect total and extravascular body water estimates, but that a fluid shift from the extracellular to the intracellular space may accompany the exchange procedure.
Cynthia J. Thornton, M.D., Donna L. Shannon, M.S., Mary Ann Hunter, R.N., Rajam S. Ramamurthy, M.D., and Yves W . Brans, M . D . S a n A n t o n i o , T e x a s
THE EFVECTS OF N E O N A T A L P O L Y C Y T H E M I A on the volumes of intravascular components (plasma, blood, and red cells) were described in an earlier study. 1 Because polycythemia and hyperviscosity may lead to sludging of red cells in various arteriolar or capillary beds, cellular hypoxia might occur regionally. This hypoxia might impair cellular metabolism and result in disturbances of cellular hydration. This study was therefore designed to identify the effects of polycythemia and of its treatment by partial exchange transfusion on water content of the various body compartments.
MATERIALS AND METHODS Ten normocythemic and eight polycythemic neonates were studied with their parents' consent. In all cases umbilical cords were clamped after the onset of breathing From the Perinatal Research Laboratory, Departments of Pediatrics and of Obstetrics and Gynecology, University of Texas Health Science Center. Supported in part by Biomedical Research Grant RR05654 of the National Institutes of Health. Reprint address: Yves W. Brans, M.D., Department of Pediatrics, University of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78284.
0022-3476/83/010113+05500.50/0
9 1983 The C. V. Mosby Co.
while the babies were held at the level of the mother's introitus, so that all neonates received a "placental transfusion." Birth weights were recorded to the nearest 10 gm. Gestational ages were calculated from the mother's menstrual history, often but not always checked by sonographic determination of biparietal diameter and confirmed by physical examination of the neonate. 2 Normal intrauterine growth for neonates of Latin-American ethnic background was defined by a birth weight between the tenth and ninetieth percentiles for gestational age (Gibbs, unpublished data). Intrauterine growth retardation was defined by a birth weight below the tenth percentile. Macrosomia was defined by a birth weight in excess of the ninetieth percentile. Polycythemia was defined by an umbilical venous hematocrit level of 63% or more? All polycythemic neonates were treated by partial exchange transfusion with Plasmanate (Cutter Laboratories, Berkeley, Calif.) as previously described? Total body water (antipyrine space), extracellular water (corrected bromide space), and plasma volume (10-minute albumin space with T-1824 as the albumin tag) were estimated simultaneously after intravenous injection via the umbilical venous catheter of a sterile solution contain-
Vol. 102, No. 1, pp. 113-117
1 14
Thornton et al.
The Journal o f Pediatrics January 1983
Table. Description of study population (mean _+ SE and range)
Characteristic
Birth weight (gin)
Normocythemic neonates (n = to)
3,020 + 186 (2,180-3,700) Gestational age (wk) 38 _+ 0.4 (37-40) Postnatal age (hr) 6.7 + 0.61 (4-i0.6) Umbilical venous hemato49 _+ 2.8 crit (%) (37-61) Umbilical venous viscosity -at 11.5 sec-~ shear rate (cps)
Polycythemic neonates (n = 8)
3,190 _+ 193 (2,440-4,160) 39 _+ 0.5 (36-40) 6.9 _+ 0.71 (4-10) 64 _+ 0.3 (63-65) 18 _+ 1.5 (14-23)
ing 0.5 gm/dl antipyrine (1-phenyl-2,3 dimethylpyrazolone-one) (Lemmon Company, Sellersville, Pa.), 3.9 gm/dl bromide (sodium salt) (Bios Coutelier, Brussels), and 22.6 mg/dl T-1824 (Evans' blue) (Harvey Laboratories, Philadelphia) in 0.9 gm/dl sodium chloride, The exact amount of marker injected was calculated from the measured concentration of the injectate, the weight of injectate administered, and the specific gravity of the injectate. An average of 10 mg/kg antipyrine (range 9 to 13 mg/kg), 67 mg/kg bromide, (range 56 to 80 mg/kg) and 0.5 mg/kg T-1824 (range 0.4 to 0.6 mg/kg) were injected. Antipyrine concentrations were determined in triplicate on the injecrate solution and on plasma samples obtained before injection of the marker and one and two hours after injection. A microadaptation of Mendelsohn and Levin's technique was used for assay? Zero-time concentration of antipyrine was extrapolated from the one- and two-hour postinjection concentrations, using the least-squares method for estimating the best fit of these points to a straight line. Antipyrine space was calculated as Antipyrine space = Amount of antipyrine injected X 0.934 Zero-time plasma antipyrine concentration where 0.934 is a correction factor for the proportion of water in plasma? Bromide cncentrations were determined in triplicate on the injectate solution and on plasma samples obtained prior to and one hour after injection of the marker by a microadaptation of Wolf and Eadie's technique.6 The corrected bromide space was calculated using the formula Corrected bromide space -Amount of bromide injected 0.90 • 1-Hour plasma bromide concentrations 0.92X 0.934 where 0.90 corrects for an estimated 10% intracellular
bromide] 0.95 corrects for the Donnan equilibrium, and 0.934 corrects for the proportion of water in plasma. Concentrations of T-1824 were determined in triplicate on the injectate solution and on plasma samples obtained prior to and exactly 10 minutes after injection of the marker by a double-wavelength technique/ Intracellu ar water was assumed to be the difference between antipyrine and corrected bromide space. Interstitial water was assumed to be the difference between corrected bromide space and plasma volume. All results were expressed in milliliters per kilogram body weight. In polycythemic neonates, body water estimates were obtained immediately before and immediately after partial exchange transfusion. In addition, blood samples were obtained exactly 10, 20, and 30 minutes after injection of the marker. Transcapillary escape rate of T-1824 was calculated from the decreasing dye concentrations in the samples by linear regression analysis. Blood viscosity was measured in a Wells-Brookfield microviscometer equipped with spindle CP-7? Results from normocythemic and polycythemic neonates were compared by means of the Student t test for two means. In polycythemic neonates, pre- and postexchange results were compared by the Student paired t test. RESULTS Ten normocythemic and eight polycythemic neonates were studied (Table). The two groups were similar in birth weight, gestational age, and postnatal age. Birth weights ranged from 2,180 to 4,160 gm; four neonates weighed < 2,500 gm, and only one weighed > 4,000 gm. Gestational ages ranged from 36 to 40 weeks, only one polycythemic neonate being less than 36 weeks. All but two neonates showed normal growth: one normocythemic baby was intrauterine growth retarded, and one polycythemic baby was macrosomic. Postnatal ages at the beginning of the study ranged from 4 to 10.6 hours. Umbilical venous hematocrits ranged from 37 to 61% in normocythemic and from 63 to 65% in polycythemic neonates. Umbilical venous viscosity at a shear rate of 11.5 sec -l, measured in only five polycythemic neonates, ranged from 14 to 23 cps, indicating that four babies had hyperviscous blood and one had a borderline value? There were no statistically significant differences between preexchange polycythemic and normocythemic neonates in mean (_+SE) total body water (742 _+ 21.4 ml/kg vs 722 _+ 24.5 ml/kg), mean extracellular water (520 + 33.9 ml/kg vs 510 _+ 21.5 ml/kg), mean intracellular water (223 _+ 32.1 ml/kg vs 212 + 29.4 ml/kg), and mean interstitial water (477 _ 34 ml/kg vs 460 _+ 21.4 ml/kg) (Fig. 1). In the polycythemic group, partial exchange transfusion with an average of 12 _+ 0.3 ml/kg
Volume 102 Number I
Body water estimates in neonatal polycythemia
I000
700
r
900 v
115
o
600
E
E ft..) 8 0 0
r
o
,,,,11,--
500
o
o
~ 13
0
700 @
133
000
9
o
o
o
600
500
o
4O0 l l
-'x W
300
200 Normocythemie
Polycythemie Pre Post
Normocythemic
Polycythemic Post
Pre
700 -
500 -
600-
c~ 4 0 0
E
9 o
3oo
I
9
" E
o
,.- 5 0 0 (1)
o 4=
o
_.o 2 0 0 -
.i
400
o o o
o c
o
9
G,) .II-
c-
IO0-
0
300
200 Normocythemic
Polycythemic Pre
Post
Normocythemic
Polycythemic Pre
Post
Fig. 1. Body water estimates in normocythemic neonates ( o ) and in poIycythemic neonates before (e) and after (a) partial exchange transfusion. Mean _+ 2 SE superimposed on individual values.
Plasmanate did not affect mean total body water (preexchange 742 _+ 21.4 m l / k g vs postexchange 727 _+ 27.6 ml/kg). It did, however, decrease mean extracellular water (520 _+ 33.9 m l / k g vs 405 ___ 21.5 ml/kg, P = 0.002) and mean interstitial water (447 _+ 34 m l / k g vs 348 + 22.1 ml/kg, P = 0.001), and increased mean intracellular
water (223 _+ 32.1 m l / k g vs 321 + 31.3 ml/kg, P = 0.019). Mean intracellular water was also increased when expressed as a proportion of total body water (ICW/TBW 0.300 _+ 0.0394 vs 0,437 + 0.0346, P = 0.005). Partial exchange transfusion did not have an apprecia-
116
Thornton et at.
The Journal o f Pediatrics January 1983
50 O
B
n--
40 co
Ld
>,T
o
9~ . -.--ID CO tO r ID
:50 B
20
lib
10
Pre
Post
Fig. 2. Transcapillary escape.rate of T-1824 in polycythemic neonates before (e) and after (A) partial exchange transfusion. Mean _+ 2 SE superimposed on individual values. ble effect on mean T-1824 transcapillary escape rate (preexchange 35 + 3%/hr vs postexehange 30 _ 4.9%/ hr) (Fig. 2). Individual values decreased in three infants, increased in two, and remained approximately the same in two; transcapillary exchange rate was not determined in one infant. DISCUSSION In an earlier paper we reported the effects of partial exchange transfusion on the various intravascular volumes. t In normally grown neonates, plasma volumes increased, but not as much as expected from the volume exchanged; red cell volume decreased approPriately; and blood volume did not change significantly. Thus a loss of plasma occurred during or after the procedure. The high transcapillary escape rates of albumin observed in this study after partial exchange transfusion (mean 35%/hr; range 15 to 44%/hr) were compatible with a rapid disappearance of plasma after the exchange. The reasons for these high transcapillary escape rates cannot be determined from the data, but several possibilities may be identified: increased hydrostatic pressure in the capillaries, decreased osmotic pressure, increased capillary permeability, and increased surface area of perfused capillaries. Increased hydrostatic pressure was not likely in the presence of relatively similar blood volumes. Decreased osmotic pressure may not be ruled out, although the use of a plasmalike fluid to replace red cells should preclude'such occurrence. Increased surface area of perfused capillaries could result from decreased viscosity and improved blood flow, but there was no evidence that blood flow was
impaired in the first place. Increased capillary permeability could be caused by hypoxic damage to the endothelial cells. The results of our investigation indicate that the fluid changes occurring with partial exchange transfusion are not limited to the vascular space, and that moderate polycythemia (umbilical venous hematocrit level 63 to 65%) does not affect total and extravascular water volumes in term neonates with essentially normal growth. In polycythemic neonates, partial exchange transfusion with a plasma equivalent did not affect total body water content but appeared to produce a shift of water from the extracellular to the intracellular space. Whether the situation thereby produced was akin to an at least transient "sick cell syndrome," as has been reported in intrauterine growth-retarded neonates,8 cannot be determined without data on blood flow or oxygen delivery to the tissues. Whatever the reason for the water shift, it must be emphasized that the lack of sequential data does not permit us to speculate on how long this abnormal state might last nor on how much damage, if any, it might cause. Transcapillary escape rates before partial exchange transfusion averaged 30%/hr (range 23 to 45%/hr). This was higher than published average values for normal neonates, which range from 13 to 31%/hr. 9,~~ Increased hydrostatic pressure is the most likely explanation, because polycythemic neonates have higher blood volumes than those of their normocythemic peers. ~Blood flow apparently was not impaired by moderate polycythemia and hyperviscosity. These data raise the possibility that partial exchange transfusion over a 10- to 15-minute period might not be the ideal way of treating neonatal polycythemia, and might in fact be harmful. More definitive conclusions may not be drawn from the available data. More data are needed to define better which polycythemic neonates need to be treated and which treatment modality is most appropriate to minimize the type of abnormalities elicited in this study. Controlled manipulations needed to obtain these data might better be performed in an animal model. REFERENCES 1. BransYW, Shannon DL, Ramamurthy RS: Neonatal polycythemia. I1. Plasma, blood, and red cell volume estimates in relation to hematocrit levels and quality of intrauterine growth. Pediatrics 68:175, 1981. 2. Dubowitz LMS, Dubowitz V, Goldberg C: Clinical assessment of gestational age in the newborn infant. J PEDIATR 77:1, 1970. 3. Ramamurthy RS, Brans YW: Neonatal polycythemia. I. Criteria for diagnosis and treatment. Pediatrics 68:168, 1981. 4. Mendelsohn D, Levin NW: A colorimetric micromethod for
Volume 102 Number 1
Clinical and laboratory observations
the estimation of antipyrine in plasma and serum. S Aft J Med Sci 25:13, I960. 5. Cassady G: Bromide space studies in infants of low birthweight. Pediatr Res 4:20, 1970. 6. Wolf RL, Eadie GS: Reabsorption of bromide by the Kidney. Am J Physiol 163:436, 1939. 7. Cheek DB, Talbert JL: Extracellular water (and sodium) and body water in infants. In Cheek DB, editor: Human growth: Body composition, cell growth, energy and intelligence. Philadelphia, 1968, Lea & Febiger, pp 117-134.
117
8. Cassady G, Milstead RR: Antipyrine space studies and cell water estimates in infants of low birthweight. Pediatr Res 5:563, 1971. 9. Cassady G: Plasma volume studies in low birthweight i'nfants. Pediatrics 38:1020, 1966. 10. lngomar CJ, Klebe JG, Baekgaard P: The transcapillary escape rate of T-1824 in healthy newborn infants: The influence of placental transfusion. Acta Paediatr Scand 62:617, 1973.
Clinical and laboratory observations Breast-feeding, weight loss, and jaundice M. Jeffrey Maisels, M.B., B.Ch., and Kathleen Gifford, R.N.C. Hershey, Pa.
INITIATION AND MAINTENANCE of successful lactation requires that a mother has the desire to breast-feed and that informed and enthusiastic professional support is available to her. Apart from potential problems associated with the technique of nursing, there are three conditions that, not infrequently, concern both mother and physician: excessive weight loss, "dehydration fever," and jaundice. Surprisingly little information is available on the normal weight loss to be expected for fully breast-fed infants, particularly when babies are divided into weight groups. Two recent texts dealing with the subject of breast-feeding fail to provide any data on this subject?' 2 Furthermore, it is unclear whether the fever that occurs occasionally in breast-fed infants is related to undernutrition or to dehydration. 3 If "breast milk fever" is in fact related to poor intake, documentation of excessive weight loss would permit the problem to be anticipated, thus avoiding unnecessary investigations and, on occasion, admission to a neonatal intensive care unit, We investigated the weight loss experienced by fully breast-fed infants and its association with jaundice and fever. From the Division of Newborn Medicine, Department of Pediatrics, The Milton S. Hershey Medical Center, The Pennsylvania State University. Reprint address: M. Jeffrey Maisels, M.D., Department of Pediatrics, The Milton S. Hershey Medical Center, Hershey, PA 17033.
0022-3476/83/010117+02500.20/09
1983 The C. V. Mosby Co.
Table. Cumulative weight loss by day 3 and total serum bilirubin concentrations Birth weight (gin)
Cumulative weight loss Day 3 (%)
250I to 3000 3001 to 3500 3501 to 4000 > 4000
5.4 5.9 6.0 6.0
+ 3.3 -+ 3.3 -+ 3.2 _+ 3.0
Serum bilirubin Day 3 (me/d# 7.5 7.0 6.2 7.1
+ 3.5 _+ 3.4 _+ 3.6 _+ 3.7
Values are means _+ SD.
METHODS W e reviewed the charts of 100 infants from the wellbaby nursery of The Milton S. Hershey Medical Center. Consecutive charts were examined until 25 infants in each of four weight groups had been identified: birthweight 2,501 to 3,000 gm, 3,001 to 3,500 gm, 3,501 to 4,000 gm, and > 4,000 gin. Infants were selected only if they had been delivered vaginally, had been fully breast-fed, and had received no supplementation with water or formula. All infants nursed from their mothers soon after birth and were breast-fed subsequently on demand. Changes in weight were charted over the first three days of life. Because most infants had been discharged by the fourth day, the numbers were too small to provide longitudinal data beyond day 3. Axillary temperatures were obtained every eight hours
113
Body water estimates in neonatal polycythemia To determine whether neonatal polycythemia and its treatment by partial exchange transfusion affect body water estimates, 10 normocythemie and eight polycythemic neonates were studied within 12 hours of birth. Total body water, extracellular water, and plasma volume were estimated immediately prior to and following exchange, lntracellular and interstitial water contenr were calculated. There were no significant differences between normocythemic and preexchange polycythemic neonates in mean total body water, extracellular water, interstitial water, and intracellular water contents. In the polycythemic group, exchange did not affect mean total body water, but was associated with decreases in mean extracellular water and mean interstitial water and an increase in mean intraeellular water. Mean transeapillary escape rate of T-1824 was not affected by exchange but was quite rapid both before (35 + SE 3%/hr) and after the procedure (30 +_ 4.9%/hr). These data suggest that moderate polyeythemia in normal term neonates does not affect total and extravascular body water estimates, but that a fluid shift from the extracellular to the intracellular space may accompany the exchange procedure.
Cynthia J. Thornton, M.D., Donna L. Shannon, M.S., Mary Ann Hunter, R.N., Rajam S. Ramamurthy, M.D., and Yves W . Brans, M . D . S a n A n t o n i o , T e x a s
THE EFVECTS OF N E O N A T A L P O L Y C Y T H E M I A on the volumes of intravascular components (plasma, blood, and red cells) were described in an earlier study. 1 Because polycythemia and hyperviscosity may lead to sludging of red cells in various arteriolar or capillary beds, cellular hypoxia might occur regionally. This hypoxia might impair cellular metabolism and result in disturbances of cellular hydration. This study was therefore designed to identify the effects of polycythemia and of its treatment by partial exchange transfusion on water content of the various body compartments.
MATERIALS AND METHODS Ten normocythemic and eight polycythemic neonates were studied with their parents' consent. In all cases umbilical cords were clamped after the onset of breathing From the Perinatal Research Laboratory, Departments of Pediatrics and of Obstetrics and Gynecology, University of Texas Health Science Center. Supported in part by Biomedical Research Grant RR05654 of the National Institutes of Health. Reprint address: Yves W. Brans, M.D., Department of Pediatrics, University of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78284.
0022-3476/83/010113+05500.50/0
9 1983 The C. V. Mosby Co.
while the babies were held at the level of the mother's introitus, so that all neonates received a "placental transfusion." Birth weights were recorded to the nearest 10 gm. Gestational ages were calculated from the mother's menstrual history, often but not always checked by sonographic determination of biparietal diameter and confirmed by physical examination of the neonate. 2 Normal intrauterine growth for neonates of Latin-American ethnic background was defined by a birth weight between the tenth and ninetieth percentiles for gestational age (Gibbs, unpublished data). Intrauterine growth retardation was defined by a birth weight below the tenth percentile. Macrosomia was defined by a birth weight in excess of the ninetieth percentile. Polycythemia was defined by an umbilical venous hematocrit level of 63% or more? All polycythemic neonates were treated by partial exchange transfusion with Plasmanate (Cutter Laboratories, Berkeley, Calif.) as previously described? Total body water (antipyrine space), extracellular water (corrected bromide space), and plasma volume (10-minute albumin space with T-1824 as the albumin tag) were estimated simultaneously after intravenous injection via the umbilical venous catheter of a sterile solution contain-
Vol. 102, No. 1, pp. 113-117
1 14
Thornton et al.
The Journal o f Pediatrics January 1983
Table. Description of study population (mean _+ SE and range)
Characteristic
Birth weight (gin)
Normocythemic neonates (n = to)
3,020 + 186 (2,180-3,700) Gestational age (wk) 38 _+ 0.4 (37-40) Postnatal age (hr) 6.7 + 0.61 (4-i0.6) Umbilical venous hemato49 _+ 2.8 crit (%) (37-61) Umbilical venous viscosity -at 11.5 sec-~ shear rate (cps)
Polycythemic neonates (n = 8)
3,190 _+ 193 (2,440-4,160) 39 _+ 0.5 (36-40) 6.9 _+ 0.71 (4-10) 64 _+ 0.3 (63-65) 18 _+ 1.5 (14-23)
ing 0.5 gm/dl antipyrine (1-phenyl-2,3 dimethylpyrazolone-one) (Lemmon Company, Sellersville, Pa.), 3.9 gm/dl bromide (sodium salt) (Bios Coutelier, Brussels), and 22.6 mg/dl T-1824 (Evans' blue) (Harvey Laboratories, Philadelphia) in 0.9 gm/dl sodium chloride, The exact amount of marker injected was calculated from the measured concentration of the injectate, the weight of injectate administered, and the specific gravity of the injectate. An average of 10 mg/kg antipyrine (range 9 to 13 mg/kg), 67 mg/kg bromide, (range 56 to 80 mg/kg) and 0.5 mg/kg T-1824 (range 0.4 to 0.6 mg/kg) were injected. Antipyrine concentrations were determined in triplicate on the injecrate solution and on plasma samples obtained before injection of the marker and one and two hours after injection. A microadaptation of Mendelsohn and Levin's technique was used for assay? Zero-time concentration of antipyrine was extrapolated from the one- and two-hour postinjection concentrations, using the least-squares method for estimating the best fit of these points to a straight line. Antipyrine space was calculated as Antipyrine space = Amount of antipyrine injected X 0.934 Zero-time plasma antipyrine concentration where 0.934 is a correction factor for the proportion of water in plasma? Bromide cncentrations were determined in triplicate on the injectate solution and on plasma samples obtained prior to and one hour after injection of the marker by a microadaptation of Wolf and Eadie's technique.6 The corrected bromide space was calculated using the formula Corrected bromide space -Amount of bromide injected 0.90 • 1-Hour plasma bromide concentrations 0.92X 0.934 where 0.90 corrects for an estimated 10% intracellular
bromide] 0.95 corrects for the Donnan equilibrium, and 0.934 corrects for the proportion of water in plasma. Concentrations of T-1824 were determined in triplicate on the injectate solution and on plasma samples obtained prior to and exactly 10 minutes after injection of the marker by a double-wavelength technique/ Intracellu ar water was assumed to be the difference between antipyrine and corrected bromide space. Interstitial water was assumed to be the difference between corrected bromide space and plasma volume. All results were expressed in milliliters per kilogram body weight. In polycythemic neonates, body water estimates were obtained immediately before and immediately after partial exchange transfusion. In addition, blood samples were obtained exactly 10, 20, and 30 minutes after injection of the marker. Transcapillary escape rate of T-1824 was calculated from the decreasing dye concentrations in the samples by linear regression analysis. Blood viscosity was measured in a Wells-Brookfield microviscometer equipped with spindle CP-7? Results from normocythemic and polycythemic neonates were compared by means of the Student t test for two means. In polycythemic neonates, pre- and postexchange results were compared by the Student paired t test. RESULTS Ten normocythemic and eight polycythemic neonates were studied (Table). The two groups were similar in birth weight, gestational age, and postnatal age. Birth weights ranged from 2,180 to 4,160 gm; four neonates weighed < 2,500 gm, and only one weighed > 4,000 gm. Gestational ages ranged from 36 to 40 weeks, only one polycythemic neonate being less than 36 weeks. All but two neonates showed normal growth: one normocythemic baby was intrauterine growth retarded, and one polycythemic baby was macrosomic. Postnatal ages at the beginning of the study ranged from 4 to 10.6 hours. Umbilical venous hematocrits ranged from 37 to 61% in normocythemic and from 63 to 65% in polycythemic neonates. Umbilical venous viscosity at a shear rate of 11.5 sec -l, measured in only five polycythemic neonates, ranged from 14 to 23 cps, indicating that four babies had hyperviscous blood and one had a borderline value? There were no statistically significant differences between preexchange polycythemic and normocythemic neonates in mean (_+SE) total body water (742 _+ 21.4 ml/kg vs 722 _+ 24.5 ml/kg), mean extracellular water (520 + 33.9 ml/kg vs 510 _+ 21.5 ml/kg), mean intracellular water (223 _+ 32.1 ml/kg vs 212 + 29.4 ml/kg), and mean interstitial water (477 _ 34 ml/kg vs 460 _+ 21.4 ml/kg) (Fig. 1). In the polycythemic group, partial exchange transfusion with an average of 12 _+ 0.3 ml/kg
Volume 102 Number I
Body water estimates in neonatal polycythemia
I000
700
r
900 v
115
o
600
E
E ft..) 8 0 0
r
o
,,,,11,--
500
o
o
~ 13
0
700 @
133
000
9
o
o
o
600
500
o
4O0 l l
-'x W
300
200 Normocythemie
Polycythemie Pre Post
Normocythemic
Polycythemic Post
Pre
700 -
500 -
600-
c~ 4 0 0
E
9 o
3oo
I
9
" E
o
,.- 5 0 0 (1)
o 4=
o
_.o 2 0 0 -
.i
400
o o o
o c
o
9
G,) .II-
c-
IO0-
0
300
200 Normocythemic
Polycythemic Pre
Post
Normocythemic
Polycythemic Pre
Post
Fig. 1. Body water estimates in normocythemic neonates ( o ) and in poIycythemic neonates before (e) and after (a) partial exchange transfusion. Mean _+ 2 SE superimposed on individual values.
Plasmanate did not affect mean total body water (preexchange 742 _+ 21.4 m l / k g vs postexchange 727 _+ 27.6 ml/kg). It did, however, decrease mean extracellular water (520 _+ 33.9 m l / k g vs 405 ___ 21.5 ml/kg, P = 0.002) and mean interstitial water (447 _+ 34 m l / k g vs 348 + 22.1 ml/kg, P = 0.001), and increased mean intracellular
water (223 _+ 32.1 m l / k g vs 321 + 31.3 ml/kg, P = 0.019). Mean intracellular water was also increased when expressed as a proportion of total body water (ICW/TBW 0.300 _+ 0.0394 vs 0,437 + 0.0346, P = 0.005). Partial exchange transfusion did not have an apprecia-
116
Thornton et at.
The Journal o f Pediatrics January 1983
50 O
B
n--
40 co
Ld
>,T
o
9~ . -.--ID CO tO r ID
:50 B
20
lib
10
Pre
Post
Fig. 2. Transcapillary escape.rate of T-1824 in polycythemic neonates before (e) and after (A) partial exchange transfusion. Mean _+ 2 SE superimposed on individual values. ble effect on mean T-1824 transcapillary escape rate (preexchange 35 + 3%/hr vs postexehange 30 _ 4.9%/ hr) (Fig. 2). Individual values decreased in three infants, increased in two, and remained approximately the same in two; transcapillary exchange rate was not determined in one infant. DISCUSSION In an earlier paper we reported the effects of partial exchange transfusion on the various intravascular volumes. t In normally grown neonates, plasma volumes increased, but not as much as expected from the volume exchanged; red cell volume decreased approPriately; and blood volume did not change significantly. Thus a loss of plasma occurred during or after the procedure. The high transcapillary escape rates of albumin observed in this study after partial exchange transfusion (mean 35%/hr; range 15 to 44%/hr) were compatible with a rapid disappearance of plasma after the exchange. The reasons for these high transcapillary escape rates cannot be determined from the data, but several possibilities may be identified: increased hydrostatic pressure in the capillaries, decreased osmotic pressure, increased capillary permeability, and increased surface area of perfused capillaries. Increased hydrostatic pressure was not likely in the presence of relatively similar blood volumes. Decreased osmotic pressure may not be ruled out, although the use of a plasmalike fluid to replace red cells should preclude'such occurrence. Increased surface area of perfused capillaries could result from decreased viscosity and improved blood flow, but there was no evidence that blood flow was
impaired in the first place. Increased capillary permeability could be caused by hypoxic damage to the endothelial cells. The results of our investigation indicate that the fluid changes occurring with partial exchange transfusion are not limited to the vascular space, and that moderate polycythemia (umbilical venous hematocrit level 63 to 65%) does not affect total and extravascular water volumes in term neonates with essentially normal growth. In polycythemic neonates, partial exchange transfusion with a plasma equivalent did not affect total body water content but appeared to produce a shift of water from the extracellular to the intracellular space. Whether the situation thereby produced was akin to an at least transient "sick cell syndrome," as has been reported in intrauterine growth-retarded neonates,8 cannot be determined without data on blood flow or oxygen delivery to the tissues. Whatever the reason for the water shift, it must be emphasized that the lack of sequential data does not permit us to speculate on how long this abnormal state might last nor on how much damage, if any, it might cause. Transcapillary escape rates before partial exchange transfusion averaged 30%/hr (range 23 to 45%/hr). This was higher than published average values for normal neonates, which range from 13 to 31%/hr. 9,~~ Increased hydrostatic pressure is the most likely explanation, because polycythemic neonates have higher blood volumes than those of their normocythemic peers. ~Blood flow apparently was not impaired by moderate polycythemia and hyperviscosity. These data raise the possibility that partial exchange transfusion over a 10- to 15-minute period might not be the ideal way of treating neonatal polycythemia, and might in fact be harmful. More definitive conclusions may not be drawn from the available data. More data are needed to define better which polycythemic neonates need to be treated and which treatment modality is most appropriate to minimize the type of abnormalities elicited in this study. Controlled manipulations needed to obtain these data might better be performed in an animal model. REFERENCES 1. BransYW, Shannon DL, Ramamurthy RS: Neonatal polycythemia. I1. Plasma, blood, and red cell volume estimates in relation to hematocrit levels and quality of intrauterine growth. Pediatrics 68:175, 1981. 2. Dubowitz LMS, Dubowitz V, Goldberg C: Clinical assessment of gestational age in the newborn infant. J PEDIATR 77:1, 1970. 3. Ramamurthy RS, Brans YW: Neonatal polycythemia. I. Criteria for diagnosis and treatment. Pediatrics 68:168, 1981. 4. Mendelsohn D, Levin NW: A colorimetric micromethod for
Volume 102 Number 1
Clinical and laboratory observations
the estimation of antipyrine in plasma and serum. S Aft J Med Sci 25:13, I960. 5. Cassady G: Bromide space studies in infants of low birthweight. Pediatr Res 4:20, 1970. 6. Wolf RL, Eadie GS: Reabsorption of bromide by the Kidney. Am J Physiol 163:436, 1939. 7. Cheek DB, Talbert JL: Extracellular water (and sodium) and body water in infants. In Cheek DB, editor: Human growth: Body composition, cell growth, energy and intelligence. Philadelphia, 1968, Lea & Febiger, pp 117-134.
117
8. Cassady G, Milstead RR: Antipyrine space studies and cell water estimates in infants of low birthweight. Pediatr Res 5:563, 1971. 9. Cassady G: Plasma volume studies in low birthweight i'nfants. Pediatrics 38:1020, 1966. 10. lngomar CJ, Klebe JG, Baekgaard P: The transcapillary escape rate of T-1824 in healthy newborn infants: The influence of placental transfusion. Acta Paediatr Scand 62:617, 1973.
Clinical and laboratory observations Breast-feeding, weight loss, and jaundice M. Jeffrey Maisels, M.B., B.Ch., and Kathleen Gifford, R.N.C. Hershey, Pa.
INITIATION AND MAINTENANCE of successful lactation requires that a mother has the desire to breast-feed and that informed and enthusiastic professional support is available to her. Apart from potential problems associated with the technique of nursing, there are three conditions that, not infrequently, concern both mother and physician: excessive weight loss, "dehydration fever," and jaundice. Surprisingly little information is available on the normal weight loss to be expected for fully breast-fed infants, particularly when babies are divided into weight groups. Two recent texts dealing with the subject of breast-feeding fail to provide any data on this subject?' 2 Furthermore, it is unclear whether the fever that occurs occasionally in breast-fed infants is related to undernutrition or to dehydration. 3 If "breast milk fever" is in fact related to poor intake, documentation of excessive weight loss would permit the problem to be anticipated, thus avoiding unnecessary investigations and, on occasion, admission to a neonatal intensive care unit, We investigated the weight loss experienced by fully breast-fed infants and its association with jaundice and fever. From the Division of Newborn Medicine, Department of Pediatrics, The Milton S. Hershey Medical Center, The Pennsylvania State University. Reprint address: M. Jeffrey Maisels, M.D., Department of Pediatrics, The Milton S. Hershey Medical Center, Hershey, PA 17033.
0022-3476/83/010117+02500.20/09
1983 The C. V. Mosby Co.
Table. Cumulative weight loss by day 3 and total serum bilirubin concentrations Birth weight (gin)
Cumulative weight loss Day 3 (%)
250I to 3000 3001 to 3500 3501 to 4000 > 4000
5.4 5.9 6.0 6.0
+ 3.3 -+ 3.3 -+ 3.2 _+ 3.0
Serum bilirubin Day 3 (me/d# 7.5 7.0 6.2 7.1
+ 3.5 _+ 3.4 _+ 3.6 _+ 3.7
Values are means _+ SD.
METHODS W e reviewed the charts of 100 infants from the wellbaby nursery of The Milton S. Hershey Medical Center. Consecutive charts were examined until 25 infants in each of four weight groups had been identified: birthweight 2,501 to 3,000 gm, 3,001 to 3,500 gm, 3,501 to 4,000 gm, and > 4,000 gin. Infants were selected only if they had been delivered vaginally, had been fully breast-fed, and had received no supplementation with water or formula. All infants nursed from their mothers soon after birth and were breast-fed subsequently on demand. Changes in weight were charted over the first three days of life. Because most infants had been discharged by the fourth day, the numbers were too small to provide longitudinal data beyond day 3. Axillary temperatures were obtained every eight hours